Environmental factors predictive of misbehavior in collaborative health research

描述(由申请人提供)：R21提案的项目摘要，“合作健康研究中预测不当行为的环境因素”。健康研究中的不当行为--如捏造研究数据、欺诈性账单或向研究参与者隐瞒有关已知风险的信息--对医疗保健经济产生不利影响，并破坏患者的信任，这反过来又导致健康差距和不愿登记参加临床试验。该项目建议使用历史计量学方法(如下所述)来确定在协作健康研究中预测严重不当行为的环境因素。环境因素包括保密、角色冲突、不当行为的奖励、模糊的职业规范、监督失败和处于权威地位等因素。该项目的长期目标是通过增加对影响研究中不当行为的实际因素的了解，改进研究道德教育计划、研究政策和监督实践。该项目的4个具体目标包括：[1]从合作研究中至少90个已发表的不当行为案例中提取数据，以生成案例摘要，其中包括关于关键主角、不当行为、事件结果的描述性数据，以及假设与研究不当行为相关的12个环境因素。我们将使用目的性标准和异质性抽样方法来获得病例。要被纳入，病例必须涉及1900年或更晚的美国的卫生科学研究；病例必须至少1岁(以确保有足够的报告机会)，并在至少3个独立发表的报告中进行描述。我们将进一步寻找代表广泛严重不当行为的案件，包括各种不当行为(例如，我们不会只关注欺诈或狭义不当行为案件)，以及涉及不同程度合作的案件。[2]使用相关和线性回归模型检验环境因素(预测变量)和不良行为严重程度(目标变量)之间的关系。独立的研究伦理专家将使用研究团队开发的评分表对每个案例中不当行为的严重程度进行评分；2名研究助理将使用基准评分表对环境因素的存在进行评分。[3]检验研究中不良行为的严重程度与两种协作形式之间的关系：(A)多个PI之间的多站点协作和(B)跨学科协作(涉及不同学科的关键人员的研究)。如果合作预示着不那么严重或更严重的不当行为，那么我们将检查变量是否单独运行，或者更确切地说，通过其他变量的中介作用，例如“模棱两可的职业规范”。[4]发表文章，探讨我们的发现对RCR教育、RCR政策和监督实践的实际影响。 公共卫生相关性(由申请人提供)：R21提案的项目说明，“合作卫生研究中预测不当行为的环境因素”。健康研究中的不当行为--如捏造研究数据、欺诈性账单或向研究参与者隐瞒有关已知风险的信息--对医疗保健经济产生不利影响，并破坏患者的信任，这反过来又会导致健康差距和不愿参加临床试验。该项目的长期目标是通过增加对影响研究中不当行为的实际因素的了解，改进研究道德教育计划、研究政策和监督实践。

DESCRIPTION (provided by applicant): Project Summary of the R21 proposal, "Environmental factors predictive of misbehavior in collaborative health research." Misbehavior in health research -- such as fabricating research data, fraudulent billing, or withholding information about known risks from research participants-adversely affects the health care economy and undermines patient trust, which in turn contributes to health disparities and reluctance to enroll in clinical trials. This project proposes to use historiometric methods (described below) to identify environmental factors that predict severe misbehavior in collaborative health research. Environmental factors include factors such as secrecy, conflicting roles, rewards for misbehavior, ambiguous professional norms, failures of oversight, and being in a position of authority. The long-term aim of this project is to improve research ethics education programs, research policies, and oversight practices by increasing understanding of the actual factors that influence misbehavior in research. The 4 specific aims of the project include: [1] To extract data from at least 90 published cases of misbehavior in collaborative research to generate case summaries that include descriptive data on the key protagonist, the misbehavior, the outcomes of the event, and 12 environmental factors hypothesized to correlate with research misbehavior. We will use purposive criterion and heterogeneity sampling methods to obtain cases. To be included, cases must involve health science research in the US from 1900 or later; cases must be at least 1 year old (to ensure adequate reporting opportunity) and be described in at least 3 independent published accounts. We will further seek out cases representing a broad spectrum of severity of misbehavior, with a variety of kinds of misbehavior represented (e.g., we will not focus just on fraud or narrow misconduct cases), and cases involving differing levels of collaboration. [2] To use correlation and linear regression models to test the relationship of environmental factors (predictor variables) and the severity of misbehavior (target variable). Independent research ethics experts will score the severity of misbehavior in each case using a score sheet developed by the research team; 2 research assistants will score the presence of environmental factors using a benchmark score sheet. [3] To test the relationship between the severity of misbehavior in research to 2 forms of collaboration: (a) multisite collaboration among multiple PIs and (b) interdisciplinary collaboration (studies involving key personnel from different disciplines). If collaboration is predictive of less severe or more severe misbehavior, then we will examine whether the variable operates in isolation or rather through the mediation of other variables such as "ambiguous professional norms." [4] To publish articles that explore the practical implications of our findings for RCR education, RCR policies, and oversight practices. PUBLIC HEALTH RELEVANCE (provided by applicant): Project Narrative for the R21 proposal, "Environmental factors predictive of misbehavior in collaborative health research." Misbehavior in health research-such as fabricating research data, fraudulent billing, or withholding information about known risks from research participants-adversely affects the health care economy and undermines patient trust, which in turn contributes to health disparities and reluctance to enroll in clinical trials. The long-term aim of this project is to improve research ethics education programs, research policies, and oversight practices by increasing understanding of the actual factors that influence misbehavior in research.